In this paper we present two contributions for a set of local transformations (a selection set) to improve a performance of a very large circuit. The first contribution is an idea of "padding node" and "multi-separator-set. " We have proven that combination of padding node and multi-separator-set provides the optimum selection set. The second contribution is our heuristic method to find a semi-optimum multi-separator-set, which uses a network flow algorithm. Our method is robust for very large circuits, because its memory usage and calculation time are linear and polynomial order with the size of the circuit. We have compared our method with Singh's selection function method, which provides the optimum selection set and is the best method in literature to date. Our method has successfully optimized delays of all circuits, while Singh's selection function method has aborted with three large circuits because of memory overflow. The results also has shown our method has a comparable capability in delay optimization to Singh's method, although our method is heuristic.

Solutions based on error-correcting codes for the blacklisting problem of a broadcast distribution system have been proposed by Kumar, Rajagopalan and Sahai. In this paper, detailed analysis of the solutions is presented. By choosing parameters properly in their constructions, we show that the performance is improved significantly.

This paper presents a novel image coding scheme based on separate coding of region and residue sources. In a subband image coding scheme, quantization errors in each subimage spread over the reconstructed image and result in a blurring or a boundary artifact. To obtain high compression ratio without considerable degradation, an input image, in our scheme, is separated into region and residue sources which are coded using different coding schemes. The region source is coded by adaptive arithmetic coder. The residue source is coded using multiresolution subimages generated by applying a subband filter. Each block in the subimages is predicted by an affine transformation of blocks in lower resolution subimages. Experimental results show that a high coding efficiency is achieved using the proposed scheme, especially in terms of the subjective visual quality and PSNR at low bit-rate compression.

We propose a new intelligent blind source separation algorithm for the mixture of sub-Gaussian and super-Gaussian sources. The algorithm consists of an update equation of the separating matrix and an adjustment equation of nonlinear functions. To verify the validity of the proposed algorithm, we compare the proposed algorithm with extant methods.

We propose a blind source separation algorithm for the mixture of finite alphabet sources where sensors are less than sources. The algorithm consists of an update equation of an estimated mixing matrix and enumeration of the inferred sources. We present the bound of a step size for the stability of the algorithm and two methods of assignment of the initial point of the estimated mixing matrix. Simulation results verify the proposed algorithm.

We have developed spectral separation devices for processing femtosecond pulses. These devices are based on an optical coupler structure with fiber gratings. In a computer simulation, we confirmed that these devices could extract <1 nm bandwidth light with 80% efficiency. We fabricated the spectral separation devices using single mode fibers and highly Ge-doped fibers. These devices successfully extracted narrow spectral light of 0.3 nm bandwidth with 37% efficiency from femtosecond pulses of 40 nm bandwidth. We also fabricated 2-channel spectral separation devices, which could extract the light from each grating channel.

Dislocation properties in InGaN/GaN Quantum Wells and GaN grown on bulk GaN and sapphire substrates by metalorganic chemical vapor deposition (MOCVD) were characterized using cathodoluminescnece (CL), transmission electron microscopy (TEM), atomic force microscopy (AFM) and photoluminescence (PL). It was clearly demonstrated that dislocations act as nonradiative recombination centers in both n-type (undoped and Si-doped) GaN and InGaN layers. Furthermore the very short-minority carrier diffusion length was a key parameter to explain the high light emission efficiency in GaN-based light emitting diodes (LEDs) prepared on sapphire substrates. On the other side band-tail states were detected in the heteroepitaxial InGaN layers only by temperature dependence PL measurement. Additionally InGaN phase separation, which consists of few micron domains, has been produced under growth conditions which favors the spiral growth. These results indicate that the dislocations in the InGaN layers act as triggering centers for the InGaN phase separation which cause both a compositional fluctuation and the formation of few micron phase separated domains. The homoepitaxial InGaN layers showed however quite normal behaviors for all characterizations.

This paper surveys how geometric information can be effectively used for efficient algorithms with focus on clustering problems. Given a complete weighted graph G of n vertices, is there a partition of the vertex set into k disjoint subsets so that the maximum weight of an innercluster edge (whose two endpoints both belong to the same subset) is minimized? This problem is known to be NP-complete even for k = 3. The case of k=2, that is, bipartition problem is solvable in polynomial time. On the other hand, in geometric setting where vertices are points in the plane and weights of edges equal the distances between corresponding points, the same problem is solvable in polynomial time even for k 3 as far as k is a fixed constant. For the case k=2, effective use of geometric property of an optimal solution leads to considerable improvement on the computational complexity. Other related topics are also discussed.

In this paper, we discuss a neural network approach for blind signal extraction of temporally correlated sources. Assuming autoregressive models of source signals, we propose a very simple neural network model and an efficient on-line adaptive algorithm that extract, from linear mixtures, a temporally correlated source with an arbitrary distribution, including a colored Gaussian source and a source with extremely low value (or even zero) of kurtosis. We then combine these extraction processing units with deflation processing units to extract such sources sequentially in a cascade fashion. Theory and simulations show that the proposed neural network successfully extracts all arbitrarily distributed, but temporally correlated source signals from linear mixtures.

We report relations between invariant manifolds of saddle orbits (Lyapunov family) around a saddle-center equilibrium point and lowest periodic orbits on the two degree of freedom swing equation system. The system consists of two generators operating onto an infinite bus. In this system, a stable equilibrium point represents the normal operation state, and to understand its basin structure is important in connection with practical situations. The Lyapunov families appear under conservative conditions and their invariant manifolds constitute separatrices between trapped and divergent motions. These separatrices continuously deform and become basin boundaries, if changing the system to dissipative one, so that to investigate those manifolds is meaningful. While, in the field of two degree of freedom motions, systems with saddle loops to a saddle-center are well studied, and existence of transverse homoclinic structure of separatrix manifolds is reported. However our investigating system has no such loops. It is interesting what separatrix structure exists without trivial saddle loops. In this report, we focus on above invariant manifolds and lowest periodic orbits which are foliated for the Hamiltonian level.

This paper develops an algorithm based on the Modular Approach to solve singly constrained separable discrete optimization problems (Nonlinear Knapsack Problems). The Modular Approach uses fathoming and integration techniques repeatedly. The fathoming reduces the decision space of variables. The integration reduces the number of variables in the problem by combining several variables into one variable. Computational experiments for "hard" test problems with up to 1000 variables are provided. Each variable has up to 1000 integer values.

In this paper, a new feature which is characterized by the extrema density of 2-D wavelet frames estimated at the output of the corresponding filter bank is proposed for texture segmentation. With and without feature selection, the discrimination ability of features based on pyramidal and tree-structured decompositions are comparatively studied using the extrema density, energy, and entropy as features, respectively. These comparisons are demonstrated with separable and non-separable wavelets. With the three-, four-, and five-category textured images from Brodatz album, it is observed that most performances with feature selection improve significantly than those without feature selection. In addition, the experimental results show that the extrema density-based measure performs best among the three types of features investigated. A Min-Min method based on genetic algorithms, which is a novel approach with the spatial separation criterion (SPC) as the evaluation function is presented to evaluate the segmentation performance of each subset of selected features. In this work, the SPC is defined as the Euclidean distance within class divided by the Euclidean distance between classes in the spatial domain. It is shown that with feature selection the tree-structured wavelet decomposition based on non-separable wavelet frames has better performances than the tree-structured wavelet decomposition based on separable wavelet frames and pyramidal decomposition based on separable and non-separable wavelet frames in the experiments. Finally, we compare to the segmentation results evaluated with the templates of the textured images and verify the effectiveness of the proposed criterion. Moreover, it is proved that the discriminatory characteristics of features do spread over all subbands from the feature selection vector.

For given undirected graph G[V,E] and vertices s and t, a minimal s-t separating set denoted by Ec & Vc is a minimal set of elements (edges and/or vertices) such that deletion of the elements from G breaks all the paths between s and t, where Ec and Vc are sets of edges and vertices, respectively. In this paper, we consider a problem of generating all minimal s-t separating sets, and show that the problem can be solved in O(µ(mt(n,n))) time, where m|E|, n|V|, µ is the number of minimal s-t separating sets of G, and t(p,q) is the time needed for finding q lowest common ancestors for q pairs of vertices in a rooted tree with p vertices. Since t(n,n) can be O(n), we can generate all minimal s-t separating in linear time per s-t separating set. However, the linear time algorithm for finding the lowest common ancestors is complicated, so that it is not efficient for a moderate size graph. Therefore, we use an O(nα (n))-time algorithm for finding the lowest common ancestors, and propose an algorithm to generate all minimal s-t separating sets in O(mnα(n)) time per s-t separating set, where α(n) is the pseudo-inverse of Ackermann function.

Although there are many multimedia presentation systems on the market, they have some shortcomings and most of them only can work on one single computer, and few of them can work on Web. Thus, in the thesis we develop a network multimedia presentation system to let users easily design the multimedia presentation without restriction on technology or presentation time and place. Our system includes 3 main components: User Interface that includes temporal specification editor, spatial specification editor and multimedia object interface, Presentation Interface and Knowledge Base. There is a dynamic homepage generator in our system and we propose a displaying algorithm based on the Allen's theory, that there exist 13 temporal relationships between two intervals, for synchronizing the media objects.

We present a cascade neural network for blind source extraction. We propose a family of unconstrained optimization criteria, from which we derive a learning rule that can extract a single source signal from a linear mixture of source signals. To prevent the newly extracted source signal from being extracted again in the next processing unit, we propose another unconstrained optimization criterion that uses knowledge of this signal. From this criterion, we then derive a learning rule that deflates from the mixture the newly extracted signal. By virtue of blind extraction and deflation processing, the presented cascade neural network can cope with a practical case where the number of mixed signals is equal to or larger than the number of sources, with the number of sources not known in advance. We prove analytically that the proposed criteria both for blind extraction and deflation processing have no spurious equilibria. In addition, the proposed criteria do not require whitening of mixed signals. We also demonstrate the validity and performance of the presented neural network by computer simulation experiments.

In this paper, the performances of texture classification based on pyramidal and uniform decomposition are comparatively studied with and without feature selection. This comparison using the subband variance as feature explores the dependence among features. It is shown that the main problem when employing 2-D non-separable wavelet transforms for texture classification is the determination of the suitable features that yields the best classification results. A Max-Max algorithm which is a novel evaluation function based on genetic algorithms is presented to evaluate the classification performance of each subset of selected features. It is shown that the performance with feature selection in which only about half of features are selected is comparable to that without feature selection. Moreover, the discriminatory characteristics of texture spread more in low-pass bands and the features extracted from the pyramidal decomposition are more representative than those from the uniform decomposition. Experimental results have verified the selectivity of the proposed approach and its texture capturing characteristics.

Unique characteristics such as quenching of photoluminescence and improvement of photovoltaic effect were observed in acceptor polymer, (cyano-substituted poly (p-phenylene vinylene)), CN-PPV/donor polymer (poly(3-hexylthiophene), P3HT composites. By taking account of the difference in electronic energy states of both CN-PPV and P3HT, these characteristics are interpreted in terms of photoinduced charge transfer between CN-PPV and P3HT and formation of fractal network.

In this paper we newly define a generalized edge-ranking of a graph G as follows: for a positive integer c, a c-edge-ranking of G is a labeling (ranking) of the edges of G with integers such that, for any label i, deletion of all edges with labels >i leaves connected components, each having at most c edges with label i. The problem of finding an optimal c-edge-ranking of G, that is, a c-edge-ranking using the minimum number of ranks, has applications in scheduling the manufacture of complex multi-part products; it is equivalent to finding a c-edge-separator tree of G having the minimum height. We present an algorithm to find an optimal c-edge-ranking of a given tree T for any positive integer c in time O(n2log Δ), where n is the number of vertices in T and Δ is the maximum vertex-degree of T. Our algorithm is faster than the best algorithm known for the case c=1.

In this paper we develop a new family of on-line adaptive learning algorithms for blind separation of time delayed and convolved sources. The algorithms are derived for feedforward and fully connected feedback (recurrent) neural networks on basis of modified natural gradient approach. The proposed algorithms can be considered as generalization and extension of existing algorithms for instantaneous mixture of unknown source signals. Preliminary computer simulations confirm validity and high performance of the proposed algorithms.

This paper presents a method to analyze and minimize output errors of 2-D non-separable FIR filters with finite wordlength. Finiteness in the wordlength causes output errors, which can be analyzed in the frequency domain when the statistics of input signals are known. The output errors can be minimized by optimizing responses corresponding to all levels of input impulses. A new ROM-based filter structure is proposed in which the optimized impulse responses are stored in the ROM. The output signals are generated by superposing the impulse responses corresponding to the input levels. Many simulation results confirm that the output signals of the proposed filters have far less errors compared to conventional filters. The hardware size of the ROM-based filters is estimated and compared with that of conventional structures. The proposed structures are more effective than the conventional ones especially when the signal wordlength is short.